Epilepsy Information

(Abst. 1.356)

Pattern of autistic-spectrum symptoms in children and adolescents with epilepsy and in neurologically-normal controls

Authors: Charles Zaroff, R. Grayson, J. Higgins, G. Vazquez and M. Lancman

Autism is associated with higher than expected rates of epilepsy/electrographic abnormality. However, specific autistic symptoms that might distinguish between those with and without epilepsy are not well reported. The purpose of the current study is to determine the specific pattern of autistic symptom presentation in a pediatric population with, and without, epilepsy, controlling for intellect and global symptom severity.


A retrospective analysis of pediatric patients referred for neuropsychological assessment was performed. Scores on the Gilliam Autism Rating Scale (GARS) and Gilliam Asperger Disorder Scale (GADS), standardized behavioral rating scales used to aid in autistic-spectrum disorder diagnoses, completed by patient caregivers, were examined in a consecutive series of 10 subjects with epilepsy and 10 subjects without epilepsy for each scale (N = 37; three epilepsy subjects were used in both GARS and GADS samples; Mean age = 9.9 years), in whom use of such measures was indicated due to caregiver complaints (e.g., socialization difficulties, restricted range of interests, communication abnormalities). Across the four sub-groups, subjects were matched for overall composite scores (GARS Autism Quotient; GADS Asperger’s Disorder Quotient) and intellectual functioning. The pattern of subtest scores on the GARS (Stereotyped Behaviors, Communication, Social Interaction, Developmental) and GADS (Social Interaction, Restricted Patterns of Behavior, Cognitive Patterns, Pragmatic Skills) was compared between the two groups. T-tests were used to compared subtest scores between the epilepsy and control group, while a one way analysis of variance (ANOVA) was used to investigate potentially significant differences between subtest scores within each group.

The mean GARS Autism Quotient was 85 in the epilepsy sample and 84 in the non-epilepsy sample, indicative of a “Below Average” Probability of Autism. The mean GADS Asperger’s Disorder Quotient was 74 in the epilepsy sample and 79 in the non-epilepsy sample, indicative of a “Borderline” Probability of Asperger’s Disorder. No statistically significant differences between the two groups on specific rating scale subtests were observed (p > .05), and a one way ANOVA revealed no significant differences between subtest scores in either group on either scale.

In subjects who are matched for overall degree of autistic symptom severity, no identifiable pattern of behavior anomaly discriminated between a pediatric sample with and without epilepsy. The negative results likely reflect in part the degree of similarity of overall global symptom severity between the two groups, a small sample size, and perhaps to a relatively mild overall degree of symptoms. Larger sample sizes and direct assessment instruments are needed. However, in the current sample, epilepsy is implicated as a co-occurring condition in autism, and not as a factor associated with exacerbation of a specific underlying autistic trait.